Age at menopause has been proposed as a marker of aging and health, but few population-based studies have been conducted addressing factors associated with age at menopause. The purpose of the current studies is to assess associations between specific medical, reproductive, behavioral, and environmental factors and menopausal status, based either on menstrual cycle patterns or follicle stimulating hormone (FSH), using population-based epidemiologic designs. We assessed associations with menopausal status using data from the National Health and Nutrition Examination Survey (NHANES) III and from the 1999 National Health Interview Survey (NHIS). We also examined the association between plasma organochlorine levels of polychlorinated biphenyls (PCBs), 1,1-dichloro-2,2-bis(p-chlorophenyl)ethylene (DDE), and age at natural menopause among 1407 women in a population-based case-control study of breast cancer. Using the NHANES III data, higher body mass index ( 30+ kg/m-squared) was associated with a lower likelihood of elevated FSH (odds ratio, OR, 0.6, 95% confidence interval, CI, 0.4, 0.9) but this association was not seen with menstrual-based menopause. Exercise (3+ times per week) was associated with a lower likelihood of being postmenopausal using menstrual (OR 0.3, 95% CI 0.2, 0.7) and hormonal (OR 0.6, 95% CI 0.4, 1.0) measures. Alcohol use also tended to be associated with postmenopausal status by either measure, but not significantly so. There was little evidence of associations with ethnicity, education, age at menarche, number of live births, and oral contraceptive use. Menstrual-based definitions of menopause can be misclassified for women with menstrual irregularity. Similar results were seen with the NHIS survey data, but education level was inversely associated with post-menopausal status (OR 0.5, 95% CI 0.3, 0.7 among women with a college degree compared with women who had not completed high school). In the study of organochlorines, the adjusted hazard ratio (HR) estimating rate of onset of natural menopause was 1.4 (95% confidence interval, CI, 0.9, 2.1) for the top decile of DDE compared with values below the median. This association is similar in magnitude to the association between smoking and menopause (HR 1.4, 95% CI 1.1, 1.9). No association was seen with PCBs.